I'm very happy to announce that the next version of the RDKit --
2012.09 (a.k.a Q3 2012) -- is released.
The release notes are below.
The source release is on the sourceforge downloads page:
The files can also be downloaded from the google project page:
The binaries for Windows, Python 2.6 and Python 2.7, have also been uploaded.
The online version of the documentation at rdkit.org has been updated.
Thanks to the everyone who submitted bug reports and suggestions for
Please let me know if you find any problems with the release or have
suggestions for the next one.
****** Release_2012.09.1 *******
(Changes relative to Release_2012.06.1)
!!!!!! IMPORTANT !!!!!!
- Some of the bug fixes affect the generation of SMILES. Canonical
SMILES generated with this version of the RDKit will be different
from previous versions.
- The fix to Issue 252 (see below) will lead to changes in calculated
logP and MR values for some compounds.
- The fix to Issue 254 (see below) will lead to changes in some
descriptors and geometries for sulfur-containing compounds.
- The fix to Issue 256 (see below) has changed the name of the
optional argument to mol.GetNumAtoms from onlyHeavy to
onlyExplicit. For compatibility reasons, Python code that uses
explicitly uses onlyHeavy will still work, but it will generate
warnings. This compatibility will be removed in a future release.
Gianpaolo Bravi, David Cosgrove, Andrew Dalke, Fabian Dey, James
Davidson, JP Ebejer, Gabriele Menna, Stephan Reiling, Roger Sayle,
- The molecules that come from mergeQueryHs() now reset the RingInfo
structure. (issue 245)
- The output from MurckoScaffold.MakeScaffoldGeneric no longer
includes stereochemistry or explicit Hs. (issue 246)
- D and T atoms in CTABs now have their isotope information
set. (issue 247)
- Some problems with ring finding in large, complex molecules have
been fixed. (issue 249)
- The "rootedAtAtom" argument for FindAllSubgraphsOfLengthN is now
handled properly. (issue 250)
- Bonds now have a SetProp() method available in Python. (issue 251)
- A number of problems with the Crippen atom parameters have been
fixed. (issue 252)
- Ring closure digits are no longer repeated on the same atom in
SMILES generated by the RDKit. (issue 253)
- Non-ring sulfur atoms adjacent to aromatic atoms are no longer set
to be SP2 hybridized. This allows them to be stereogenic. (issue
- The combineMols() function now clears computed properties on the
- A couple of problems with the pickling functions on big endian
hardware were fixed.
- The molecule drawing code now uses isotope information
- Superscript/Subscript handling in the agg canvas has been improved.
- SKP lines in CTABS are now propertly handled. (Issue 255)
- The name of the optional argument to mol.GetNumAtoms has been
changed from onlyHeavy to onlyExplicit. The method counts the number
of atoms in the molecular graph, not the number of heavy
atoms. These numbers happen to usually be the same (which is why
this has taken so long to show up), but there are exceptions if Hs
or dummy atoms are in the graph. (Issue 256)
- Unknown bonds in SMILES are now output using '~' instead of '?'. The
SMILES parser now recognizes '~' as an "any bond" query. (Issue 257)
- Lines containing only white space in SDF property blocks are no
longer treated as field separators.
- Transition metals and lanthanides no longer have default valences
- The RDKit now has a maximum common substructure (MCS) implementation
contributed by Andrew Dalke. This is currently implemented in Python
and is available as: from rdkit.Chem import MCS Documentation is
available as a docstring for the function MCS.FindMCS and in the
- A few new functions have been added to rdkit.Chem.Draw:
MolsToImage(), MolsToGridImage(), ReactionToImage()
- CalcMolFormula() now provides the option to include isotope
- The RDKit and Layered fingerprinters both now accept "fromAtoms"
arguments that can be used to limit which atoms contribute to the
- Version information is now available in the Java wrapper.
- The descriptor NumRadicalElectrons is now available.
- The PyMol interface now supports a GetPNG() method which returns the
current contents of the viewer window as an PIL Image object.
- Molecules (ROMol in C++, rdkit.Chem.Mol in Python) now have a
- Component-level grouping (parens) can be used in reaction SMARTS.
New Database Cartridge Features:
- support for molecule <-> pickle conversion via the functions
mol_to_pkl, mol_from_pkl, and is_valid_mol_pkl.
- support for bit vector <-> binary text conversion via the functions
New Java Wrapper Features:
Deprecated modules (to be removed in next release):
- During this release cycle, the sourceforge project was updated to
their new hosting system. This explains the change in bug/issue
- the SMILES parser is now substantially faster.
- The molecular drawings generated by Code/Demo/RDKit/Draw/MolDrawing.h
have been improved.
- There is now demo code availble for using the C++ drawing code
within Qt applications. (contributed by David Cosgrove)
- The directory $RDBASE/Regress now contains sample data and
scripts for benchmarking the database cartridge.
- Fused-ring aromaticity is now only considered in rings of up to size
- It is no longer necessary to have flex and bison installed in order
to build the RDKit.
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
Rdkit-devel mailing list